address, but the associated PF is giving the VF an all zeros MAC address
when one is not administratively assigned. The driver should check for
this case and generate a random address, similar to how the linux igbvf
driver does.
Submitted by: skoumjian@juniper.net (Scott Koumjian)
MFH: 2 weeks
Differential Revision: https://reviews.freebsd.org/D8399
free'd by the functions following its call, we can simply return instead
of crashing and burning in the event of igb_detach() failing.
PR: 197139
Submitted by: rupavath@juniper.net
MFC after: 2 weeks
And factor out tcp_lro_rx_done, which deduplicates the same logic with
netinet/tcp_lro.c
Reviewed by: gallatin (1st version), hps, zbb, np, Dexuan Cui <decui microsoft com>
Sponsored by: Microsoft OSTC
Differential Revision: https://reviews.freebsd.org/D5725
- At Intel it is believed that most of their products support "only"
40 DMA segments so lower {EM,IGB}_MAX_SCATTER accordingly. Actually,
40 is more than plenty to handle full size TSO packets so it doesn't
make sense to further distinguish between MAC variants that really
can do 64 DMA segments. Moreover, capping at 40 DMA segments limits
the stack usage of {em,igb}_xmit() that - given the rare use of more
than these - previously hardly was justifiable, while still being
sufficient to avoid the problems seen with em(4) and EM_MAX_SCATTER
set to 32.
- In igb(4), pass the actually supported TSO parameters up the stack.
Previously, the defaults set in if_attach_internal() were applied,
i. e. a maximum of 35 TSO segments, which made supporting more than
these in the driver pointless. However, this might explain why no
problems were seen with IGB_MAX_SCATTER at 64.
- In em(4), take the 5 m_pullup(9) invocations performed by em_xmit()
in the TSO case into account when reporting TSO parameters upwards.
In the worst case, each of these calls will add another mbuf and,
thus, the requirement for an additional DMA segment. So for best
performance, it doesn't make sense to advertize a maximum of TSO
segments that typically will require defragmentation in em_xmit().
Again, this leaves enough room to handle full size TSO packets.
- Drop TSO macros from if_lem.h given that corresponding MACS don't
support TSO in the first place.
Reviewed by: erj, sbruno, jeffrey.e.pieper_intel.com
Approved by: erj
MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D5238
Major changes:
- Add i219/i219(2) hardware support. (Found on Skylake generation and newer
chipsets.)
- Further to the last Skylake support diff, this one also includes support for
the Lewisburg chipset (i219(3)).
- Add a workaround to an igb hardware errata.
All 1G server products need to have IPv6 extension header parsing turned off.
This should be listed in the specification updates for current 1G server
products, e.g. for i350 it's errata #37 in this document:
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/ethernet-controller-i350-spec-update.pdf
- Avoton (i354) PHY errata workaround added
And a bunch of minor fixes, as well as #defines for things that the current
em(4)/igb(4) drivers don't implement.
Differential Revision: https://reviews.freebsd.org/D3162
Reviewed by: sbruno, marius, gnn
Approved by: gnn
MFC after: 2 weeks
Sponsored by: Intel Corporation
- Add optimizing LRO wrapper which pre-sorts all incoming packets
according to the hash type and flowid. This prevents exhaustion of
the LRO entries due to too many connections at the same time.
Testing using a larger number of higher bandwidth TCP connections
showed that the incoming ACK packet aggregation rate increased from
~1.3:1 to almost 3:1. Another test showed that for a number of TCP
connections greater than 16 per hardware receive ring, where 8 TCP
connections was the LRO active entry limit, there was a significant
improvement in throughput due to being able to fully aggregate more
than 8 TCP stream. For very few very high bandwidth TCP streams, the
optimizing LRO wrapper will add CPU usage instead of reducing CPU
usage. This is expected. Network drivers which want to use the
optimizing LRO wrapper needs to call "tcp_lro_queue_mbuf()" instead
of "tcp_lro_rx()" and "tcp_lro_flush_all()" instead of
"tcp_lro_flush()". Further the LRO control structure must be
initialized using "tcp_lro_init_args()" passing a non-zero number
into the "lro_mbufs" argument.
- Make LRO statistics 64-bit. Previously 32-bit integers were used for
statistics which can be prone to wrap-around. Fix this while at it
and update all SYSCTL's which expose LRO statistics.
- Ensure all data is freed when destroying a LRO control structures,
especially leftover LRO entries.
- Reduce number of memory allocations needed when setting up a LRO
control structure by precomputing the total amount of memory needed.
- Add own memory allocation counter for LRO.
- Bump the FreeBSD version to force recompilation of all KLDs due to
change of the LRO control structure size.
Sponsored by: Mellanox Technologies
Reviewed by: gallatin, sbruno, rrs, gnn, transport
Tested by: Netflix
Differential Revision: https://reviews.freebsd.org/D4914
alignment guarantees provided by m_defrag(9), use m_collapse(9)
instead for performance reasons.
While at it, sanitize the statistics softc members, i. e. retire
unused ones and add SYSCTL nodes missing for actually used ones.
Differential Revision: https://reviews.freebsd.org/D4717
in igb and fix a wrap-around bug.
Reviewed by: hiren
Obtained from: Jason (j@nitrology.com)
MFC after: 2 weeks
Sponsored by: LimeLight Networks
Differential Revision: https://reviews.freebsd.org/D4039
to attach with the last version of this commit. This commit fixes
attach failures on "ICH8" class devices via modifications to
e1000_init_nvm_params_ich8lan()
- Fix compiler warning in 80003es2lan.c
- Add return value handler for e1000_*_kmrn_reg_80003es2lan
- Fix usage of DEBUGOUT
- Remove unnecessary variable initializations.
- Removed unused variables (complaints from gcc).
- Edit defines in 82571.h.
- Add workaround for igb hw errata.
- Shared code changes for Skylake/I219 support.
- Remove unused OBFF and LTR functions.
Tested by some of the folks that reported breakage in previous incarnation.
Thanks to AllanJude, gjb, gnn, tijl for tempting fate with their machines.
Submitted by: erj@freebsd.org
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D3162
This commit contains large contributions from Giuseppe Lettieri and
Stefano Garzarella, is partly supported by grants from Verisign and Cisco,
and brings in the following:
- fix zerocopy monitor ports and introduce copying monitor ports
(the latter are lower performance but give access to all traffic
in parallel with the application)
- exclusive open mode, useful to implement solutions that recover
from crashes of the main netmap client (suggested by Patrick Kelsey)
- revised memory allocator in preparation for the 'passthrough mode'
(ptnetmap) recently presented at bsdcan. ptnetmap is described in
S. Garzarella, G. Lettieri, L. Rizzo;
Virtual device passthrough for high speed VM networking,
ACM/IEEE ANCS 2015, Oakland (CA) May 2015
http://info.iet.unipi.it/~luigi/research.html
- fix rx CRC handing on ixl
- add module dependencies for netmap when building drivers as modules
- minor simplifications to device-specific routines (*txsync, *rxsync)
- general code cleanup (remove unused variables, introduce macros
to access rings and remove duplicate code,
Applications do not need to be recompiled, unless of course
they want to use the new features (monitors and exclusive open).
Those willing to try this code on stable/10 can just update the
sys/dev/netmap/*, sys/net/netmap* with the version in HEAD
and apply the small patches to individual device drivers.
MFC after: 1 month
Sponsored by: (partly) Verisign, Cisco
- Use hardware counters for ifnet stats in igb(4) when possible. This
ensures these stats include packets that bypass the regular stack via
netmap.
- Don't derefence values off the end of the igb(4) VF stats structure.
Instead, add a dedicated if_get_counter method for igb(4) VF interfaces.
- Report missed packets on igb(4) as input queue drops rather than an
input error.
- Report bug_ring drop counts as output queue drops for igb(4) and ixgbe(4).
- Export the buf_ring drop stats for individual rings via sysctl on
ixgbe(4).
- Fix a typo that in ixl(4) that caused output queue drops to be reported
as input queue drops and input queue drops to be unreported.
Differential Revision: https://reviews.freebsd.org/D2402
Reviewed by: jfv, rstone (6)
Sponsored by: Norse Corp, Inc.
RSS hash from the card. We do not need to hide that under "ifdef RSS" and should
expose that by default so others like lagg(4) can use that and avoid hashing the
traffic by themselves.
While here, improve comments and get rid of hidden/unimplemented RSS support
code for UDP.
Differential Revision: https://reviews.freebsd.org/D2296
Reviewed by: jfv, erj
Discussed with: adrian
Sponsored by: Limelight Networks
bits.
The motivation here is to eventually teach netisr and potentially
other networking subsystems a bit more about how RSS work queues / buckets
are configured so things have a hope of auto-configuring in the future.
* net/rss_config.[ch] takes care of the generic bits for doing
configuration, hash function selection, etc;
* topelitz.[ch] is now in net/ rather than netinet/;
* (and would be in libkern if it didn't directly include RSS_KEYSIZE;
that's a later thing to fix up.)
* netinet/in_rss.[ch] now just contains the IPv4 specific methods;
* and netinet/in6_rss.[ch] now just contains the IPv6 specific methods.
This should have no functional impact on anyone currently using
the RSS support.
Differential Revision: D1383
Reviewed by: gnn, jfv (intel driver bits)
from the FreeBSD network code. The flag is still kept around in the
"sys/mbuf.h" header file, but does no longer have any users. Instead
the "m_pkthdr.rsstype" field in the mbuf structure is now used to
decide the meaning of the "m_pkthdr.flowid" field. To modify the
"m_pkthdr.rsstype" field please use the existing "M_HASHTYPE_XXX"
macros as defined in the "sys/mbuf.h" header file.
This patch introduces new behaviour in the transmit direction.
Previously network drivers checked if "M_FLOWID" was set in "m_flags"
before using the "m_pkthdr.flowid" field. This check has now now been
replaced by checking if "M_HASHTYPE_GET(m)" is different from
"M_HASHTYPE_NONE". In the future more hashtypes will be added, for
example hashtypes for hardware dedicated flows.
"M_HASHTYPE_OPAQUE" indicates that the "m_pkthdr.flowid" value is
valid and has no particular type. This change removes the need for an
"if" statement in TCP transmit code checking for the presence of a
valid flowid value. The "if" statement mentioned above is now a direct
variable assignment which is then later checked by the respective
network drivers like before.
Additional notes:
- The SCTP code changes will be committed as a separate patch.
- Removal of the "M_FLOWID" flag will also be done separately.
- The FreeBSD version has been bumped.
MFC after: 1 month
Sponsored by: Mellanox Technologies
This allows one to make a kernel module to tune the
number of queues before the driver loads.
This is needed so that a module at SI_SUB_CPU can set
tunables for these drivers to take. Otherwise getenv
is called too early by the TUNABLE macros.
Reviewed by: smh
Phabric: https://reviews.freebsd.org/D1149
- Wrong integer type was specified.
- Wrong or missing "access" specifier. The "access" specifier
sometimes included the SYSCTL type, which it should not, except for
procedural SYSCTL nodes.
- Logical OR where binary OR was expected.
- Properly assert the "access" argument passed to all SYSCTL macros,
using the CTASSERT macro. This applies to both static- and dynamically
created SYSCTLs.
- Properly assert the the data type for both static and dynamic
SYSCTLs. In the case of static SYSCTLs we only assert that the data
pointed to by the SYSCTL data pointer has the correct size, hence
there is no easy way to assert types in the C language outside a
C-function.
- Rewrote some code which doesn't pass a constant "access" specifier
when creating dynamic SYSCTL nodes, which is now a requirement.
- Updated "EXAMPLES" section in SYSCTL manual page.
MFC after: 3 days
Sponsored by: Mellanox Technologies
- Read the counts of received, dropped, and transmitted management
packets and add sysctl nodes for them.
- Fix the total octets received/transmitted to read all 64 bits of
the counters.
- Add missing sysctl nodes for rlec, tncrs, fcruc, tor, and tot.
- Remove spurious spaces.
Reviewed by: Eric Joyner @ Intel
MFC after: 1 week
double-free mbufs.
Like ixgbe(4) chipsets, EOP is only set on the final descriptor
in a chain of descriptors. So, to free the whole list of descriptors,
we should free the current slot _and_ the assembled list of descriptors
that make up the fragment list.
The existing code was setting discard once it saw EOP + an error status;
it then freed all the subsequent descriptors until the next EOP. That's
totally the wrong order.
This allows the NIC to drop frames on the receive queue and not
cause the MAC to block on receiving to _any_ queue.
Tested:
igb0@pci0:5:0:0: class=0x020000 card=0x152115d9 chip=0x15218086 rev=0x01 hdr=0x00
vendor = 'Intel Corporation'
device = 'I350 Gigabit Network Connection'
class = network
subclass = ethernet
Discussed with: Eric Joyner <eric.joyner@intel.com>
MFC after: 1 week
Sponsored by: Norse Corp, Inc.
Mostly bugfixes or features developed in the past 6 months,
so this is a 10.1 candidate.
Basically no user API changes (some bugfixes in sys/net/netmap_user.h).
In detail:
1. netmap support for virtio-net, including in netmap mode.
Under bhyve and with a netmap backend [2] we reach over 1Mpps
with standard APIs (e.g. libpcap), and 5-8 Mpps in netmap mode.
2. (kernel) add support for multiple memory allocators, so we can
better partition physical and virtual interfaces giving access
to separate users. The most visible effect is one additional
argument to the various kernel functions to compute buffer
addresses. All netmap-supported drivers are affected, but changes
are mechanical and trivial
3. (kernel) simplify the prototype for *txsync() and *rxsync()
driver methods. All netmap drivers affected, changes mostly mechanical.
4. add support for netmap-monitor ports. Think of it as a mirroring
port on a physical switch: a netmap monitor port replicates traffic
present on the main port. Restrictions apply. Drive carefully.
5. if_lem.c: support for various paravirtualization features,
experimental and disabled by default.
Most of these are described in our ANCS'13 paper [1].
Paravirtualized support in netmap mode is new, and beats the
numbers in the paper by a large factor (under qemu-kvm,
we measured gues-host throughput up to 10-12 Mpps).
A lot of refactoring and additional documentation in the files
in sys/dev/netmap, but apart from #2 and #3 above, almost nothing
of this stuff is visible to other kernel parts.
Example programs in tools/tools/netmap have been updated with bugfixes
and to support more of the existing features.
This is meant to go into 10.1 so we plan an MFC before the Aug.22 deadline.
A lot of this code has been contributed by my colleagues at UNIPI,
including Giuseppe Lettieri, Vincenzo Maffione, Stefano Garzarella.
MFC after: 3 days.
The igb(4) hardware is capable of RSS hashing RX packets and doing RSS
queue selection for up to 8 queues. (I believe some hardware is limited
to 4 queues, but I haven't tested on that.)
However, even if multi-queue is enabled for igb(4), the RX path doesn't use
the RSS flowid from the received descriptor. It just uses the MSIX queue id.
This patch does a handful of things if RSS is enabled:
* Instead of using a random key at boot, fetch the RSS key from the RSS code
and program that in to the RSS redirection table.
That whole chunk of code should be double checked for endian correctness.
* Use the RSS queue mapping to CPU ID to figure out where to thread pin
the RX swi thread and the taskqueue threads for each queue.
* The software queue is now really an "RSS bucket".
* When programming the RSS indirection table, use the RSS code to
figure out which RSS bucket each slot in the indirection table maps
to.
* When transmitting, use the flowid RSS mapping if the mbuf has
an RSS aware hash. The existing method wasn't guaranteed to align
correctly with the destination RSS bucket (and thus CPU ID.)
This code warns if the number of RSS buckets isn't the same as the
automatically configured number of hardware queues. The administrator
will have to tweak one of them for better performance.
There's currently no way to re-balance the RSS indirection table after
startup. I'll worry about that later.
Additionally, it may be worthwhile to always use the full 32 bit flowid if
multi-queue is enabled. It'll make things like lagg(4) behave better with
respect to traffic distribution.
These changes prevent sysctl(8) from returning proper output,
such as:
1) no output from sysctl(8)
2) erroneously returning ENOMEM with tools like truss(1)
or uname(1)
truss: can not get etype: Cannot allocate memory
there is an environment variable which shall initialize the SYSCTL
during early boot. This works for all SYSCTL types both statically and
dynamically created ones, except for the SYSCTL NODE type and SYSCTLs
which belong to VNETs. A new flag, CTLFLAG_NOFETCH, has been added to
be used in the case a tunable sysctl has a custom initialisation
function allowing the sysctl to still be marked as a tunable. The
kernel SYSCTL API is mostly the same, with a few exceptions for some
special operations like iterating childrens of a static/extern SYSCTL
node. This operation should probably be made into a factored out
common macro, hence some device drivers use this. The reason for
changing the SYSCTL API was the need for a SYSCTL parent OID pointer
and not only the SYSCTL parent OID list pointer in order to quickly
generate the sysctl path. The motivation behind this patch is to avoid
parameter loading cludges inside the OFED driver subsystem. Instead of
adding special code to the OFED driver subsystem to post-load tunables
into dynamically created sysctls, we generalize this in the kernel.
Other changes:
- Corrected a possibly incorrect sysctl name from "hw.cbb.intr_mask"
to "hw.pcic.intr_mask".
- Removed redundant TUNABLE statements throughout the kernel.
- Some minor code rewrites in connection to removing not needed
TUNABLE statements.
- Added a missing SYSCTL_DECL().
- Wrapped two very long lines.
- Avoid malloc()/free() inside sysctl string handling, in case it is
called to initialize a sysctl from a tunable, hence malloc()/free() is
not ready when sysctls from the sysctl dataset are registered.
- Bumped FreeBSD version to indicate SYSCTL API change.
MFC after: 2 weeks
Sponsored by: Mellanox Technologies
NULL to determine if bus_dmamap_unload() or bus_dmamem_free() should be
called. Instead, check the associated bus and virtual addresses.
- Don't clear static DMA maps to NULL.
Reviewed by: jfv
Most relevant features:
- netmap emulation on any NIC, even those without native netmap support.
On the ixgbe we have measured about 4Mpps/core/queue in this mode,
which is still a lot more than with sockets/bpf.
- seamless interconnection of VALE switch, NICs and host stack.
If you disable accelerations on your NIC (say em0)
ifconfig em0 -txcsum -txcsum
you can use the VALE switch to connect the NIC and the host stack:
vale-ctl -h valeXX:em0
allowing sharing the NIC with other netmap clients.
- THE USER API HAS SLIGHTLY CHANGED (head/cur/tail pointers
instead of pointers/count as before). This was unavoidable to support,
in the future, multiple threads operating on the same rings.
Netmap clients require very small source code changes to compile again.
On the plus side, the new API should be easier to understand
and the internals are a lot simpler.
The manual page has been updated extensively to reflect the current
features and give some examples.
This is the result of work of several people including Giuseppe Lettieri,
Vincenzo Maffione, Michio Honda and myself, and has been financially
supported by EU projects CHANGE and OPENLAB, from NetApp University
Research Fund, NEC, and of course the Universita` di Pisa.
